The patterning of the retina and its axonal output are critical for visual function, and the retinotectal proj- ection is a classic model of the development of two-dimensional neural maps. Retinal ganglion cells acquire positional coordinates along the dorsal-ventral (D-V) and anterior-posterior (A-P) axes, then translate these into graded expression of axon guidance molecules, which control topographic sorting in the optic tract and topographic targeting on the tectum. In the long term, we seek to understand how gradients of axonal beha- vior are generated. As a first step, we propose here a comprehensive analysis of the genes involved in the patterning of dorsal retina, which to date is poorly understood. Misexpression and dominant-negative experiments in chick and Xenopus have implicated BMP4, Tbx5, and ephrin-B2 in dorsal specification and D-V topography. However, the roles of Tbx5 and ephrin-B2 have not been tested by loss-of-function analysis, and our data show that more genes must be involved. We will analyze D-V retinal patterning using the zebrafish visual system, which is ideal for rapid and precise loss- of-function experiments to study retinal patterning and retinotectal topography. The main goals are: (1) to test the required roles of the known candidate genes bmp4, tbx5, and ephrin-B2 in specifying dorsal retinal fate and topographic projections; (2) to test the roles of new candidate genes we have identified, specifi- cally tcf7, other tbx genes, and ephrin-B1\ and (3) to identify new genes with a forward-genetic screen for mutants that disrupt dorsal retinal specification. Our preliminary data implicate Wnt signaling for the first time in D-V retinal patterning, and furthermore suggest crosstalk between the Wnt and BMP pathways. Building on our previous expertise in the zebrafish visual system, we will combine genetic experiments with the rapid and sophisticated embryological and phenotypic analysis possible in zebrafish. We will use ex- isting mutants and antisense morpholino oligonucleotides, as well as new mutants isolated by forward gene- tic and reverse genetic (TILLing) genetic screens. We will critically test the existing model of D-V retinal patterning, and extend it significantly by testing new hypotheses arising from our preliminary data, and by conducting an unbiased screen for required genes. These studies will address two fundamental questions in developmental neurobiology: How do regions of the nervous system acquire positional coordinates? and, How do they then make topographic connections with each other? Identifying the key genes in dorsal retina will lay the groundwork for future work on how these genes interact to generate topographic gradients. Relevance to public health. The development and structure of the visual system are remarkably conserv- ed across evolution from fish to humans. Therefore, these studies will shed light on how the human eye is formed and becomes patterned. This is of particular interest because defects in dorsoventral eye patterning are associated with human developmental disorders such as coloboma.